The specific name of the American Three-toed Woodpecker (2 Viewers)

[jmorlan]

When the AOU split the Three-toed Woodpecker, they adopted the scientific name Picoides dorsalis (Baird) for the new American Three-toed Woodpecker following Zink.

Zink et al. (The Condor 104:167–170, 2002) say, "...the first three-toed
woodpecker from North America was described as Picoides dorsalis by Baird
in 1858 (see AOU 1957). Thus, the American Three toed Woodpecker should be
reclassified as Picoides dorsalis and the Eurasian species should remain P.
tridactylus."

I don't see where the 1957 AOU checklist suppresses Brehm's earlier name but Bangs (Auk 17:126-142, 1900) invalidated Brehm's name P. americanus (Brehm) and proposed instead the name P. americanus (Swainson not of Brehm) for the American populations. Why did the AOU not go with Swainson's 1837 name? Swainson's type is based on an earlier 1831 painting labeled P. tridactylus. Is that a problem?

Thanks in advance.

 

[Richard Klim]

Sounds like a mission for mb1848...

Richard

 

[RaMa]

The first idea coming to my mind is, that somebody did not agree with Bangs and decided that Brehm's name is still available and not a nomen nudum, but at the same time considered it unidentifiable. Then Swainson's name would be preoccupied and could not be used.

However, Brehm only said that Picoides americanus is bigger than the European form (it is actually smaller!). This is the whole "description"! Even if it were correct, it's perhaps debatable whether such is sufficient to make a name available.
And no, there'd be no problem otherwise with Apternus americanus Swainson 1837, made available by reference to Swainson and Richardson, Fauna Bor.-Am.(dated to 1831 but actually published in 1832).

Rainer

 

[jmorlan]

Thank you. I also questioned the idea that Brehm's name was a nomen nudum since it has a diagnosable "description." But it did not occur to me that, if it's not identifiable or refers to the wrong taxon, it would still preoccupy Swainson's name.

Interestingly, Swainson's name was used by Swarth in 1911 and apparently by the AOU in the 1910 3rd edition according to Ridgway's synonomy. However, I confirmed that Brehm's name was still used by Ridgway in 1914.

The 4th edition of the AOU check-list lumped the American and Eurasian birds but that was reversed in the 44th Supplement in which it was claimed there was no justification for this lumping. However, the 44th supplement is silent on why it did not use Swainson's or even Brehm's earlier name instead of Baird's.

Thus far, I have been unable to find a published source where this subject is discussed.

 

[mb1848]

Bangs responded to Ridgway in 1914 using P. americanus in a 1930 article in the Bulletin of the Museum of Comparative Zoology.

Bangs: Types of birds in Museum of Comparative Zoology Harvard. p.235

PICOIDES AMERICANUS BACATUS Bangs

now PICOIDES TRIDACTYLUS BACATUS Bangs
Picoides americanus bacatus Bangs, Auk, 17, 1900, p. 136.

Type.— Ko. 100,802, cT; Maine, Bangor; 25 March, 1884; E. S.
Bowler.

In the Auk (17, 1900. pp. 126-142) I wrote a review of the American three-toed woodpeckers in which I named the small form of eastern North America, bacatus. In 1914 Ridgway in the Birds of North and Middle America did not accept my name, but used americanus Brehm (Handb. Vogel Deutschl., 1831, p. 195). I have gone into the question carefully again, and refuse to change my previous opinions, except that in one point I was wrong. I called Brehm's name americanus, a nomen nudum. Brehm did give one word of description "grossere" (as compared with European forms). The bird I name P. a. bacatus is considerably smaller than the European P. tridactylus and closely related forms, and Brehm's name, therefore, cannot be applied to it, as was done by Ridgway. The only North American three-toed woodpecker that is distinctly larger than the European is P. arcticus for which I for one refuse to use Brehm's name. Dr. Hartert tells me (m litt.) that there are no North American specimens of Picoides in the Brehm Collection. Therefore, Brehm's unrecognizable diagnosis cannot be backed up by a type specimen.

As I read Brehm again, it seems to me quite certain that he gave the name only to something that he thought occurred in America, but about which he knew nothing. It seems to me his name goes out of consideration under Opinion 2 of the International Commission of Zoological Nomenclature. Some years ago Dr. Joseph Grinnell (Univ. Calif. Pub. Zool., 5, 1909, p. 217) declared that he shared with me the opinion that americanus of Brehm cannot stand as a formal name for any subspecies. I can see no reason at all for calling the small form of eastern North America anything but bacatus.

Without the slightest hesitation, I list all black and white-backed three-toed woodpeckers as subspecies of Picoides tridactylus (Linne).


I like Swainson's P. Americanus:
Apternus, Sw. Lateral ridges close to the margin. Bill rather broader than high. Feet three-toed; the two anterior nearly equal, the posterior much longer. Arctic regions, (fig. 276.) A. Americanus.† N. Zool. pl. 56. A. arcticus. N. Z. pl. 57. † It being now ascertained that this is a distinct species from the European tridactylus, another name is necessary to designate it.

On the natural history and classification of birds, Volume 2. By William Swainson
Bottom of page 306.
But if Brehm's P. americanus negate its use perhaps Article 46 or 49 could be of help. Swainson says (I think) someone else has determined that the 3-toes in America are different (not him) therefore I will rename my 3-toe in 1832 Swainson & Richardson. He lists a lectotpe in 1837. Cassin in 1864 thought that Swainson's A. americanus and Baird's P. dorsalis were the same species.

By the way the statement "the first three-toed woodpecker from North. America was described as Picoides dorsalis by Baird in 1858" is demonstrably false, Swainson's bird was from the source of the Athabasca.

 

[jmorlan]

Hi Mark,

Thanks for the additional citation from Bangs. I think this MCZ publication is very likely the reason the American birds were lumped with the Eurasian birds in the 4th edition of the AOU checklist.

Do you know of an online source for the 4th edition of the checklist? I have been able to locate both the Bangs article and the Grinnell publications you cite in the online Biodiversity Heritage Library.

Are we then in agreement that there is nothing wrong with Swainson's name P. americanus (Swainson) and the the AOU erred in adopting P. dorsalis (Baird)?

 

[mb1848]

4th ed:
http://www.archive.org/stream/checklistofnorth1931amer/checklistofnorth1931amer_djvu.txt .
I was trying to argue that Brehm's Picoides americanus was a misidentification confusing P. arcticus with P. tridactylus. But Art. 46 & 49 do not really help. My problem is Rainer's suggestion that Swainson's A. Americanus is preoccupied sounds right. One could ask the ICZN for a ruling? I am not sure.

 

[Daniel Philippe]

In 2002 Richard Banks was still chair of the AOU Checklist Committee. We asked him to comment on this thread and this is his answer:


"It took some time to figure this out, and as far as I can tell the details of the change in names have never been spelled out. However, when the AOU most recently split the three-toed woodpeckers, we knew that dorsalis was the oldest name for the American birds and used it accordingly, not having any reason to do a complete nomenclatural history. After all, that had been the case since the fourth edition of the AOU Check-list in 1931.

But the American three-toed woodpeckers were called americanus Brehm in the third edition of the Check-list (1910), in Ridgway 1914, and Cory 1919. (Contrary to something in this exchange, we never used americanus Swainson.) So the change took place between 1919 and 1931. There were Check-list Supplements in 1920 and 1923, but no changes in these woodpecker names, so the change took place between 1923 and 1931. In 1927, Forbush ( Birds of Massachusetts) still used americanus, so the change was between then and 1931.

Bangs in 1930 published a list of the type specimens of birds in the Museum of Comparative Zoology, in the Bull. of MCZ no. 50. On p. 235 he discussed the type of the subspecies bacatus that he had described in 1900. He reviewed his 1900 paper, states that he has gone into the question carefully again, and “refuse to change my previous opinion, except that in one point I was wrong. I called Brehm’s name americanus a nomen nudum. Brehm did give one word of description ‘grössere’ (as compare to European forms).” But because the bird he had named bacatus is smaller than European tridactylus, he believed that Brehm’s name could not apply. Further, “Dr. Hartert tells me (in litt.) that there are no North American specimens of Picoides in the Brehm Collection. Therefore, Brehm’s unrecognizable diagnosis cannot be backed up by a type specimen.” He continues: “As I read Brehm again, it seems to me quite certain that he gave the name only to something that he thought occurred in America, but about which he knew nothing.” He cites Opinion 2 of the International Commission on Zoological Nomenclature (which I did not look up.)

His concluding paragraph is: “Without the slightest hesitation I list all black and white-backed three toed woodpeckers as subspecies of Picoides tridactylus (Linné).”

Thus was the name americanus disposed of and the American birds lumped with the Eurasian ones. At least, that was good enough for the Check-list Committee of that time. This is summarized very briefly, and not quite accurately, on p. 389 of the fourth edition (1931) of the check-list in a summary of changes from the previous edition, but only Bangs 1900 is cited, not Bangs 1930. There was no mention of this change in the 1930 or 1931 Auk, and it was never in a Supplement.

I hope this resolves the situation to everyone’s satisfaction."


May I suggest in the future that those who want to question the actions of the AOU check with them first before posting information that may eventually prove incorrect.

 

[jmorlan]

Thanks for obtaining this summary from Richard Banks which helps fill in some of the gaps. However I believe that Banks is wrong in stating,

"when the AOU most recently split the three-toed woodpeckers, we knew that dorsalis was the oldest name for the American birds and used it accordingly."

In fact P. americanus (Brehm) is the oldest name for the American bird and it was used by the AOU through the previous editions until the lumping in the 4th edition. However the 44th Supplement states that this lumping was without justification! In my view Bangs makes a good case for suppressing Brehm's name because it is unidentifiable. But Bangs never advocated using P. dorsalis. He advocated the next available name which was P. americanus (Swainson).

In a nutshell:

• P. americanus (Brehm) - 1831 - Unidentifiable but used by the AOU through the 3rd edition.

• P. americanus (Swainson) - 1837 - Advocated by Bangs & Grinnell.

• P. dorsalis (Baird) - 1858 - Adopted by the AOU in the 44th supplement apparently following Zink et al 2002.

Unless I'm missing something, there is nothing in Banks's statement which justifies using P. dorsalis other than the claim that it is the oldest name for the American bird. It clearly is not.

 

[RaMa]

But Bangs never advocated using P. dorsalis. He advocated the next available name which was P. americanus (Swainson).

Well, no, in 1930 he made no mention of Swainson's name but advocated lumping all subspecies in P. tridactylus. Also he changed his view regarding Brehm's name in no longer treating it as a nomen nudum. As I said above

somebody did not agree with Bangs and decided that Brehm's name is still available and not a nomen nudum, but at the same time considered it unidentifiable​

and it seems this one was Bangs himself!

So my nutshell looks like this:

• P. americanus Brehm 1831 - Not a nomen nudum but unidentifiable and therefore not usable.

• P. americanus (Swainson) 1837 - Preoccupied by P. americanus, cannot be used

• P. dorsalis Baird 1858 - The next oldest name for any of the New World subspecies, to be used if the New World form is split.

Admittedly it is quite contentious if P. americanus Brehm is really an available name. But it seems this view was taken by the AOU and it explains why P. dorsalis is now used.

Rainer

 

[jmorlan]

Well, yes, it was Bangs in 1900 who advocated for Swainson's name. That view was endorsed by Grinnell in 1909: page 218 - UC Publications in Zoology Vol. 5 1909 The 1907 Alexander Alaska Expedition.

After studying over the version of the synonomy of Picoides
americanus according to Bangs (Auk, April, 1900, pp. 127-129),
I must express myself as agreeing fully with him. The name
americanus should date from Swainson and not from Brehm.
This name therefore applies to the form of interior Alaska and
Northwest Territory, previously called successively fasciatus and
alascensis. The New England bird should stand as P. a. bacatus
Bangs

It is not apparent to me that Brehm's name preoccupies that of Swainson. I have searched through the code as best I can and have yet to find such a rule. However, there may be something in "Opinion 2" of the ICZM cited by Bangs. I have not been able to find Opinion 2 on the ICZM web site or elsewhere and would love to know what it says and if it still applies today.

In any event, Bangs's 1930 decision to revert to P. tridactylus is considered to be without good reason in the AOU's 44th supplement, so I regard that as off-point in this post-split era.

 

[RaMa]

Well, yes, it was Bangs in 1900 who advocated for Swainson's name.

It escapes my understanding what the problem here is. In 1900 Bangs considered P. americanus Brehm a nomen nudum. Accordingly P. americanus Swainson was not preoccupied and could be used. But when, in 1930, he considered Brehm's name available, Swainsons name was preoccupied.

It is not apparent to me that Brehm's name preoccupies that of Swainson. I have searched through the code as best I can and have yet to find such a rule.

The case is straightforward:

57.3 Secondary homonyms.
57.3.1 Identical species-group names established for different nominal taxa and subsequently brought together in combination with the same generic name are secondary homonyms [Art. 53.3] and the junior is invalid. ​

Picoides americanus Brehm 1831
Apternus americanus Swainson 1837, independently from Brehm
When both are in Picoides, the junior is preoccupied by the senior one. Even if Brehm's name is unidentifiable, it's undoubtedly a form of Three-toed Woodpecker.

In any event, Bangs's 1930 decision to revert to P. tridactylus is considered to be without good reason in the AOU's 44th supplement, so I regard that as off-point in this post-split era.

Yes, but so we don't know which name Bangs would have used for the American species in 1930 if he had not lumped it into the Eurasian.

Rainer

 

[jmorlan]

Even if Brehm's name is unidentifiable, it's undoubtedly a form of Three-toed Woodpecker.

Do I understand the code correctly, that to invalidate a junior homonym, the original name has to apply to a different nominal taxon? That is not the case here. Brehm's description is too vague to determine what it is. All we know is that it's bigger than a Three-toed. If it clearly applied to a Black-backed Woodpecker, I would agree with you that Swainson's name would be an invalid junior homonym. But in fact Brehm's name has a long history of being applied to the American Three-toed Woodpecker which is not a different taxon from that of Swainson.

 

[RaMa]

Do I understand the code correctly, that to invalidate a junior homonym, the original name has to apply to a different nominal taxon? That is not the case here.

The Code distinguishes between "nominal taxa" and "taxonomic taxa". Both are technical terms explained in the glossary.
"Nominal taxon" approximately (but not exactly) means "available name". All the synonyms of a species, together with its valid name, are nominal species denoting but a single taxonomic species. As Swainson named his bird independently of Brehm there are two different nominal taxa for the same species having the same specific epithet (americanus). [If Brehm had preserved a type specimen, it would've been different from Swainson's type.] So both are homonyms (and at the same time synonyms, of course).
Cases like this are admittedly quite rare, most homonyms indeed belong to different "taxonomic" species. But the Code applies to this one as well.

Rainer

 

[jmorlan]

I'm still trying to get my head around this. The code invalidates junior homonyms only for "different nominal taxa." If nominal taxa means different "available names" then how can they be homonyms? Homonyms have the same name by definition.

Nominal taxon is not specifically defined in the glossary (it just refers us to "taxon,") but we find:

taxonomic taxon A taxon (e.g. family, genus, species) including whatever nominal taxa and individuals a zoologist at any time considers it to contain in his or her endeavour to define the boundaries of a zoological taxon (q.v.). A taxonomic taxon is denoted by the valid name determined from the available names of its included nominal taxa.

To me, it seems obvious from this definition that "nominal taxa" means the synonyms which have been applied to a taxonomic taxon.

I think it's a stretch to interpret 57.3.1 to apply to anything other than different "taxonomic taxa" along with the synonyms of each. The obvious intent of 57.3.1 is to prevent different species from having the same name after their genera have been lumped. That makes sense.

Since Brehm's name has had a long history of use attached to the American Three-toed Woodpecker, it cannot, or should not, be construed to apply to a different "nominal taxon" from that of Swainson.

 

[RaMa]

Nominal taxon is not specifically defined in the glossary

Glossary, p.180
nominal taxon
A concept of a taxon which is denoted by an available name (e.g. Mollusca, Diptera, Bovidae, Papilio, Homo sapiens). Each nominal taxon in the family, genus or species groups is based on a name-bearing type (although in the latter two groups such a type may not have been actually fixed).

Picoides americanus Brehm and Apternus americanus Swainson have different types and are thus different nominal species. No way out!

I think it's a stretch to interpret 57.3.1 to apply to anything other than different "taxonomic taxa" along with the synonyms of each.

No, it must be applied to something other than different "taxonomic taxa", namely "nominal taxa", as this is precisely what it says. Whether these nominal taxa belong to the same taxonomic taxon or not is irrelevant.

The obvious intent of 57.3.1 is to prevent different species from having the same name after their genera have been lumped. That makes sense.

Agreed. And it works.

Since Brehm's name has had a long history of use attached to the American Three-toed Woodpecker, it cannot, or should not, be construed to apply to a different "nominal taxon" from that of Swainson.

But then, as Brehm's name is considered available, the species should be called Picoides americanus Brehm 1831, because this one has clearly priority!?
All would be much easier if Brehm had created an undisputable nomen nudum.

Rainer

 

[jmorlan]

Thanks. But I still don't get it. Sorry to be so dense.

type, n.A term used alone, or forming part of a compound term, to denote a particular kind of specimen or taxon.

Brehm's nominal taxon has no type, neither specimen nor a description adequate for diagnosis. It therefore cannot be considered a different "nominal taxon" from that of Swainson.

The term "different" I take to have a plain language meaning.

 

[RaMa]

Brehm's nominal taxon has no type, neither specimen nor a description adequate for diagnosis.

The type is the specimen on which a name is based, whether it was preserved or not. We don't know whether Brehm has seen a specimen of what he called P. americanus, or if he based the name on hearsay (quite probable). But surely he did not have access to Swainson's specimen and so this name cannot be based on the same specimen as Swainson's. [If it were, then both names would be objective synonyms (in addition to being secondary homonyms!), and again we would have to treat Brehm as author, because of priority]. But both names (equal spelling of their specific epithets, but not the same!) are only subjective synonyms at the species rank. This is also reflected by the fact that Brehm's species name was assigned to the eastern subspecies, while Swainson's applied to the northwestern. They were never treated as identical in meaning by any author!
Sorry, but I can offer no alternative to this, except for treating P. americanus Brehm as unavailable.

Rainer

 

[jmorlan]

This all boils down to whether Brehm's name is a nominal taxon or not. Right? If it is a nominal taxon then it is a valid name and it has priority. Otherwise it's not a nominal taxon because it's unidentifiable in which case Swainson's name is not a junior homonym. If the original name is not a nominal taxon it cannot have any junior homonyms.

In logic, this is similar to the principle of the excluded middle. One or the other has to be true but not both.

Basically I think it's a really bad idea to use an invalid name to suppress a valid one. If we allowed that, I could publish thousands of pages of gibberish binomials, each with useless one-word descriptions. My list of unidentifiable taxa would prevent the use of any of my names for newly described taxa. In fact, any use of my garbage names would have to give me credit. Consider this an example of taxonomic "denial of service."

Either Brehm's name is good which is fine with me. Or Brehm's name is no good (does not qualify because it's not a nominal taxon) in which case Swainson's is the next available by priority.

Incidentally, the article by Bangs says that Hartert reported that there were no "larger" specimens from "Amerika" or anywhere else in the Brehm collection so the name cannot be linked to a physical specimen.

 

[RaMa]

This all boils down to whether Brehm's name is a nominal taxon or not. Right?

Yes, exactly!

If it is a nominal taxon then it is a valid name and it has priority. Otherwise it's not a nominal taxon because it's unidentifiable in which case Swainson's name is not a junior homonym.

Perhaps you're confused by the difference between availability and validity? A nominal taxon is available, which means nothing more than that it's a name proposed in accordance with the Code and which can be used in zoological nomenclature, either as a valid name or as a synonym.
If P. americanus Brehm is considered a nomen nudum, it is not available and can be ignored altogether.
However, it is apparently considered available by virtually all authorities because of the description, which is one of the requirements for availability (publication is another one), but at the same time invalid, because the description allows no identification.
But to "enter into homonymy", as the Code says, a name only needs to be available, not valid!

Basically I think it's a really bad idea to use an invalid name to suppress a valid one.

The Code has a problem with unidentifiable names, yes, but it is difficult to decide where to draw a line. At least names with no description at all are excluded, but how many words of description are sufficient? Ten? Hundred? Thousand? Many old species descriptions were fully diagnosable at the time, but with the discovery of more and more new species, the differences between them became more and more subtle. Even descriptions of several pages may turn out unidentifiable at some future time, if the mentioned characters are uninformative.

If we allowed that, I could publish thousands of pages of gibberish binomials, each with useless one-word descriptions.

Actually, and most unfortunately, there are people out there doing just this, the most notorious of them botching around in herpetology and coleopterology. They are called "taxonomic vandals" and have been the cause of heated discussions in and around ICZN, without conclusive results. The names so created may number in the hundreds at least.

Rainer